(_a_) 22 18-1/2 mm. Medium thick (_b_) 22-1/2 18 mm. Medium thick

THREEPENCE.

(_a_) 22 (full) 18 mm. Very thin wove (_a_) 22 18mm. Very thin laid (_b_) 22-1/2 17-1/2 mm. Thin (_c_) 22-3/4 17-1/2 mm. Thick

SIXPENCE.

(_a_) 22 18 mm. Thin wove (_a_) 22 18 mm. Thin Laid (_b_) 22-3/4 17-3/4 mm. Thick

SEVENPENCE-HALFPENNY.

(_a_) 22-1/4 18-1/2 mm. (bare) Med. thick (_a_) 22-1/2 18-1/2 mm. Medium thick (_a_) 22-3/4 18 mm. Medium thick (_a_) 22-3/4 18-1/2 mm. Medium thick

TENPENCE.

(_a_) 22-3/4 17-1/2 mm. Thin to very thin (_b_) 22-1/2 18 mm. (full) Thick (_b_) 22-3/4 18-1/2 mm. (bare) Thick (_c_) 22 18 mm. Thin

"The varieties of the Tenpence are those described by Mr. Brouse as (_a_) long and narrow, (_b_) long and broad, and (_c_) short and broad. I may add that in the case of this value I have examined and measured some forty copies, including a strip of three, as also a proof on very thin India paper, which corresponds exactly in measurement with variety (_b_) on the thick paper (22-3/4 18-1/2mm.). It is obvious that to be absolutely accurate beyond a half mm. with an ordinary gauge is hardly possible, but in several of the given cases I have averaged the sizes of several that very closely approximated.

"As will be seen, I have gone somewhat beyond the lines of Mr.

Brouse's paper in including the 1/2d., the 3d., and 6d., the variation in the former being slight, but in the two latter noteworthy. The question how these varieties have arisen is an interesting one, nor can I see that they can be accounted for by shrinkage of the paper, as in the case of the 10d. proof above cited, which is on all fours with the ordinary stamp on thick paper.

In the case of the strip of this value I found all three stamps measured the same, and the fact remains that variety (_c_) is short _and_ broad. In any case the existence of these varieties is palpable, the question of their origin a genuine philatelic problem, and I think that the thanks of us all are therefore due to Mr.

Brouse for his interesting paper."

This may have been the first record of the peculiarity in the case of the Canadian stamps, but it was at least not the first time that variation in the dimensions of certain line engraved stamps, supposed to have been produced from the same original die, had been noted and discussed. We refer to the case of the early Ceylon stamps, which furnished food for contention in the philatelic press for many years.

The first mention of a difference in the length of these seems to have been in December, 1864.[39] Ten years later the reference list of Ceylon prepared by the London Philatelic Society[40] noted the fact that the stamps of 1863 on unwatermarked paper were in general about a millimeter shorter in the vertical dimension than the succeeding issue on paper watermarked Crown C C, although the engraved designs were otherwise absolutely identical. Major Edw. B. Evans, in his catalogue,[41] appends a note on the unwatermarked stamps of 1863 as follows:--

These stamps are apparently (indeed, we may say certainly) from the same plates as the other issues, but at the same time the impressions on this paper are about 1-16 inch shorter than those on other papers. This can only have been occasioned by the paper having shrunk to some extent since the stamps were printed....

[39] The Stamp Collectors' Magazine, II: 191.

[40] The Philatelist, IX: 10.

[41] A Catalogue for Collectors, page 39.

Later, in 1887, Mr. T. K. Tapling, writing in _Le Timbre-Poste_,[42]

claims the difference cannot be due to shrinkage of paper because the stamps have all shrunk evenly, and attributes it to some defect in the process of making the plates. He reasons thus:--

Les timbres sur les feuilles de n'importe quelle valeur etaient tous identiques comme type. Ils furent graves sur acier, je pense par MM.

Perkins Bacon et Co., chaque timbre par un procede de reduplication, etant reproduit d'une matrice; la planche etant ensuite durcie pour l'impression. Il n'y a par consequent pas de variete de types, les lignes des gravures sur les timbres courts etant les memes que celles sur les timbres longs, excepte qu'elles sont un tant soit peu contractees.... Il me semble plus que probable que la difference en longeur des exemplaires puisse etre attribuee a un leger defaut dans le procede de reduplication des planches de la matrice originale.

[42] =Le Timbre-Poste=, Numero Jubilaire, page x.x.xV.

As a matter of fact the stamps did not shrink evenly, but very unevenly.

Mr. W. B. Thornhill, writing on these same stamps in 1889,[43]

says:--"You can hardly find two stamps of exactly the same measurements in the same value, though the difference in many cases is too small to signify"; and he proceeds to show the extreme variations in a carefully prepared table including every value on every variety of paper for issues from 1855 to 1867. The greatest variation in the vertical dimension seems to be about 1 mm. in 26 mm., or roughly 4%, and in the horizontal dimension about 1/4 to 1/2 mm. in 19 mm. or roughly 1-1/4 to 2-1/2%. These dimensional differences being so palpably existent, therefore, what factors are we to consider in looking for their cause?

There seem to be but three: first, an original die or matrix for each different size; second, one original die only, whose impressions on the printing plate show variations resulting from the process of transferring them; third, a printing plate with all the impressions exact duplicates of the one original die, but whose reproductions in ink on dampened paper are varied by the shrinkage of the paper in drying.

[43] =Philatelic Record=, XI: 71.

Mr. Thornhill convinces himself by inspection that the first proposition is untenable; in fact its absurdity is at once apparent on a little thought, for the engraving of the original die is a laborious and costly piece of work, and that very fact, coupled with the comparative ease of exact reduplication by mechanical processes on the printing plate, furnishes the chief reason for the employment of this method of producing stamps. Since there is such a variety in the size of the stamps, therefore, the first theory would indicate many original dies, and this we know was not the case. Its refutation indeed is seen in the stamps themselves; for each original die, if differing in size from its fellows, meant a separate engraving, and it is humanly impossible to make these separate engravings exact duplicates, whereas, on the other hand, no appreciable variation in line or dot can be detected on the same stamp in its different sizes save the general expansion or contraction of the design, which is proportionate in all its parts. The different die or matrix theory is therefore thrown out on grounds of impracticability and absurdity.

Accepting the one original die proposition, then, Mr. Thornhill agrees with Mr. Tapling in turning down the shrinkage of paper theory and favoring the second supposition, that the variation comes on the plates and is due to the process of transference. Let us glance at this a moment. The original die is engraved on a block of soft steel of very fine and even quality. When finished it is tempered to a very great degree of hardness. Next the engraving is transferred by tremendous pressure to a transferring roller of similar soft steel, which is in turn hardened. In this process there might be an opportunity for a slight variation in the size of the transferred impression, due to the expansion and contraction of the steel in the tempering process. Next, this hardened transfer roller is impressed upon the printing plate of soft steel as many times as there are copies desired. These naturally all agree among themselves and with the transfer roller impression in size. Now when the printing plate in turn receives its hardening, there may again be a chance for a slight difference between the transfer roller and the plate impressions; _but_ it is wholly unlikely that the plate impressions will vary much among themselves, otherwise the perfection of Mr. Jacob Perkins' invention, the chief merit of which was exact reduplication, would be impaired. As a matter of fact, the high grade and even quality of the steel necessarily employed, and the care naturally taken in hardening the plate, preclude any other than an even variation, if any, due to the tempering process. This means that such variations would be practically constant over the printing surface of the plate, and that therefore the impressions would still remain practically identical in size.

Where, then, does this bring us? With such numerous and well defined variations in dimensions in the printed stamps, we should look for the cause in the simplest and most natural method by which they could readily be produced, which is furnished by the third theory presented.

Concerning this we quote from the London Philatelic Society's work on Ceylon:[44]--

In reference to the variations in the size of the stamps of Issues III and V [no watermark and Crown CC], Major Evans, who was the first to propound the theory that these variations were due to differences in the nature of the paper employed, writes as follows:--

"The theory of the expansion and contraction of the paper being now pretty generally accepted, as accounting for the variations observed in the size of the stamps of the early issues of Ceylon, it seems necessary to explain exactly what that theory is, and how these differences are supposed to arise. Previous to printing from plates engraved in _taille-douce_ the paper is wetted, which, as is well known, causes it to expand; the amount of expansion varies, no doubt, considerably in different kinds of paper, and it must also vary with the amount of moisture in the same kind of paper, for as the paper dries it returns to its original dimensions, and, therefore, up to a certain point, the wetter it is the greater will be the expansion. In any case the paper is in a state of expansion at the time of printing, both from being wetted and from being stretched out flat and pressed, and the impression when first printed is then, and then only, in all cases the size of the engraving upon the plate. It then dries, and in so doing contracts, and the greater the amount of expansion the greater will be the amount of the subsequent contraction, so that the smallest stamps are those printed on the paper which expanded most, and the largest those on the paper which expanded least. The minor variations of size may be due to the paper being more or less damp when used, but probably a very slight difference in the thickness or density of the paper would cause some variation in its expansion. The marked difference in size of the stamps on thin, unwatermarked paper, which were the first to attract the attention of Philatelists, is no doubt due to that particular variety of paper, which is very tough and elastic, and which has been found to expand very greatly on being wetted and stretched."

[44] =Postage Stamps, &c., of British India and Ceylon=, page 69.

So much for the Ceylon stamps, which we have discussed _in extenso_; but we have only to subst.i.tute in every case a reference to the first Canadian issues, particularly the 10d. which we started out with, to make the discussion apply with equal force in this case as in the other.

The question is the same--the variations occur in the same way, the method of engraving and reproduction is the same, and the varieties in the paper are very similar.

Major Evans, in a reply to Mr. Thornhill's paper,[45] states that he tried some experiments in wetting a thin, tough note paper, and found an expansion of three per cent., while by stretching it he increased the expansion to eight per cent, without difficulty! Yet the greatest variation in Mr. Thornhill's table was only four per cent. Major Evans then tried some of the 1863 Newfoundland stamps, which he judged were on paper of almost the same nature as that of the unwatermarked Ceylons of the same year, and they gave precisely similar results.

[45] =Philatelic Record=, XI: 158.

Mr. Frank C. Young, who was in the printing business, also tells of similar experiments which he carried still further.[46]

Having provided some twenty-five sheets of paper of different qualities and thicknesses, each was cut into sixteen pieces.

Selecting a common half tone cut which measured exactly 100 69 mm.

and dampening the sheets of paper to different degrees of wetness I proceeded to impress the cut on each sheet, using a common roller proof press. After the printed sheets had been allowed to dry it became a matter of a good millimeter gauge and careful measurements of the printed impressions, not the paper.

... Hardly two sheets of the whole lot were identical in size, nor was I able to formulate any table as to how much or how little or which way of the paper shrinkage would occur. The only general rule which seemed to come out clearly was that thin paper would invariably shrink more than thick. In many of the sheets the difference was barely noticeable, while, on the other hand, such measurements as 96 68, 97 68-1/2, 99 67-1/2, 98 68 mm. were fairly common, and one sheet, after several very careful measurements, was undeniably 95-1/2 69 mm., thus showing a shrinkage of 4-1/2 per cent, one way and none at all the other. This was very thin laid linen paper.

Contrary to all expectations, more than one impression measured more than either the cut or those printed on dry paper, one on thin wove paper being fully 101 mm. long.

[46] =Canada Stamp Sheet=, IV: 173.

Looking back now at Mr. Castle's tables,[47] we find his greatest variations in length amount to 3/4 mm. in 22 mm., or roughly 3-1/2%, and in width 1 mm. in 18 mm., or roughly 5-1/2%--results entirely within bounds according to Major Evans' and Mr. Young's experiments, and doubtless settling once and for all the reason of the "three distinct varieties in design" of Mr. Brouse.

[47] See page 54.

As for the paper actually used for the printing of the 10d. stamp, we find it a hard, white wove variety varying very much in thickness from a very thin, almost pelure quality, through which the design is quite plainly evident, to a medium and finally a considerably thicker quality.

The pelure paper seems naturally to be the one on which the greatest variation in dimensions occurs, the _long_ and _broad_ size of the stamp coming princ.i.p.ally on the thicker paper,[48] which is supposed to shrink the least upon drying and therefore keeps the printed impression nearest the size of the plate impression. The _long_ and _narrow_ impression, being the commoner variation, was probably due to the paper being fed to the press the same way of the "grain" as a rule, while the _short_ and _broad_ variation, which is much scarcer, occurred by an occasional sheet of paper being fed the other way of the "grain." That paper has a "grain" is readily proved by tearing a piece in one direction and then tearing it at right angles to the first tear; one will be found much easier of accomplishment generally than the other, and this "grain"

doubtless has its due effect in the amount of shrinkage in one way or the other upon drying a dampened sheet.

[48] See page 56.

One further variety we have to record in the 10d. stamp, this being a "shifted transfer" variety similar to that occurring in the 3d. value.

In this case we find the letters A D A and S of "Canada Postage," and P E N of "Pence" showing a distinct doubling at the bottom, the transfer roller evidently having been set a little too high at first and a very slight impression made on the plate. The stamp has not been seen in a pair to prove its character absolutely, but it bears all the ear-marks of being a proper plate variety and not due to a careless impression when printing.